The two-day event is aimed at promoting the Christoffel-Darboux kernel, a well known concept in approximation theory, as a versatile tool for applied areas of science, such as data analysis, dynamical systems or global optimization. The benefits of treating big data from the lens of the Christoffel-Darboux kernel are amply supported by recent advances in outlier detection, density estimation and manifold learning. Also, recent developments in non-convex, non-linear optimization single out the same kernel as an invaluable technical catalyst.

During the first day of the workshop four comprehensive lectures will introduce the non-expert participant to the theoretical and applied background of this promising, fast growing area of research.

The second day will be devoted to more specialized, seminar type discussions, including numerical experiments, explorations of real data and open questions.

Reference

The Christoffel-Darboux Kernel for Data Analysis

This volume offers foundational material for the topics covered by the workshop.

Early career researchers interested in participating with local financial support are kindly asked to send an e-mail to David Kimsey with a CV and a brief statement outlining the motivation for attending the meeting.

Registration: Registration is free, but if you intend to attend, please send an e-mail to David Kimsey.

If you are coming to the meeting and are interested in attending Monday night's dinner, please send a message to David Kimsey.